View in the dream screen.
YouTube
YouTube on Wednesday announced AI features for creators on its Shorts platform that leverage… GoogleDeepMind's video generation model.
The feature, known as Veo, lets creators add AI-generated backgrounds to their videos, as well as use written instructions to create standalone six-second videos. YouTube CEO Neil Mohan said he hopes Veo will enable creators to produce more short-form videos with the help of AI.
“Everything we showed with AI was really about enhancing the work you do, making it faster and more efficient, and bringing your creative ideas to life faster,” Mohan said, speaking at the Made on YouTube event in New York.
Veo AI Backgrounds is an upgrade to a similar AI-generated feature YouTube announced in 2023 called Dream Screen. The company said its Veo AI Background feature will roll out later this year, while the six-second AI clips will become available in 2025.
Other announcements at the event included new features in the YouTube Studio app that will let creators use AI to generate titles, thumbnails and video ideas. YouTube said these features will roll out in late 2024.
Creators have explored different ways to leverage generative AI technology. Creators have used the new technology to insert clips into their videos or produce videos that are entirely generated by AI.
However, some creators have expressed concerns about their YouTube videos being used to train the AI models that Veo is built on.
“I don’t know how to feel about all this AI stuff,” said Thomas Simmons, a comedian with more than 15 million followers on YouTube. “It doesn’t fill me with confidence and love.”
Other services like Facebook have been criticized for being overrun with AI-generated spam. There are also concerns that AI-generated content could infringe on intellectual property protections.
The company said that content generated by artificial intelligence on YouTube will be watermarked and will carry a label indicating that it was generated by artificial intelligence.
Generative AI brings a new perspective to the creator economy, giving creators free access to the tools used by large language models.
“We really sit at the intersection of this technology and creativity,” Mohan said. “Combining those two things gives us this unique lens where everything we build is really aimed at enhancing human creativity.”
